In [1]:
import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.preprocessing import StandardScaler, normalize
from sklearn.cluster import KMeans
from sklearn.decomposition import PCA
import os
import warnings
warnings.filterwarnings('ignore')
In [2]:
os.getcwd()
Out[2]:
'/Users/saikrishnaganta'
In [3]:
path = '/Users/saikrishnaganta/Acmegrade/Data Science/Projects/marketig department'
os.getcwd()
Out[3]:
'/Users/saikrishnaganta'
In [4]:
creditcard_df = pd.read_csv('Marketing_data.csv')
display (creditcard_df )
CUST_ID BALANCE BALANCE_FREQUENCY PURCHASES ONEOFF_PURCHASES INSTALLMENTS_PURCHASES CASH_ADVANCE PURCHASES_FREQUENCY ONEOFF_PURCHASES_FREQUENCY PURCHASES_INSTALLMENTS_FREQUENCY CASH_ADVANCE_FREQUENCY CASH_ADVANCE_TRX PURCHASES_TRX CREDIT_LIMIT PAYMENTS MINIMUM_PAYMENTS PRC_FULL_PAYMENT TENURE
0 C10001 40.900749 0.818182 95.40 0.00 95.40 0.000000 0.166667 0.000000 0.083333 0.000000 0 2 1000.0 201.802084 139.509787 0.000000 12
1 C10002 3202.467416 0.909091 0.00 0.00 0.00 6442.945483 0.000000 0.000000 0.000000 0.250000 4 0 7000.0 4103.032597 1072.340217 0.222222 12
2 C10003 2495.148862 1.000000 773.17 773.17 0.00 0.000000 1.000000 1.000000 0.000000 0.000000 0 12 7500.0 622.066742 627.284787 0.000000 12
3 C10004 1666.670542 0.636364 1499.00 1499.00 0.00 205.788017 0.083333 0.083333 0.000000 0.083333 1 1 7500.0 0.000000 NaN 0.000000 12
4 C10005 817.714335 1.000000 16.00 16.00 0.00 0.000000 0.083333 0.083333 0.000000 0.000000 0 1 1200.0 678.334763 244.791237 0.000000 12
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
8945 C19186 28.493517 1.000000 291.12 0.00 291.12 0.000000 1.000000 0.000000 0.833333 0.000000 0 6 1000.0 325.594462 48.886365 0.500000 6
8946 C19187 19.183215 1.000000 300.00 0.00 300.00 0.000000 1.000000 0.000000 0.833333 0.000000 0 6 1000.0 275.861322 NaN 0.000000 6
8947 C19188 23.398673 0.833333 144.40 0.00 144.40 0.000000 0.833333 0.000000 0.666667 0.000000 0 5 1000.0 81.270775 82.418369 0.250000 6
8948 C19189 13.457564 0.833333 0.00 0.00 0.00 36.558778 0.000000 0.000000 0.000000 0.166667 2 0 500.0 52.549959 55.755628 0.250000 6
8949 C19190 372.708075 0.666667 1093.25 1093.25 0.00 127.040008 0.666667 0.666667 0.000000 0.333333 2 23 1200.0 63.165404 88.288956 0.000000 6

8950 rows × 18 columns

In [5]:
creditcard_df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 8950 entries, 0 to 8949
Data columns (total 18 columns):
 #   Column                            Non-Null Count  Dtype  
---  ------                            --------------  -----  
 0   CUST_ID                           8950 non-null   object 
 1   BALANCE                           8950 non-null   float64
 2   BALANCE_FREQUENCY                 8950 non-null   float64
 3   PURCHASES                         8950 non-null   float64
 4   ONEOFF_PURCHASES                  8950 non-null   float64
 5   INSTALLMENTS_PURCHASES            8950 non-null   float64
 6   CASH_ADVANCE                      8950 non-null   float64
 7   PURCHASES_FREQUENCY               8950 non-null   float64
 8   ONEOFF_PURCHASES_FREQUENCY        8950 non-null   float64
 9   PURCHASES_INSTALLMENTS_FREQUENCY  8950 non-null   float64
 10  CASH_ADVANCE_FREQUENCY            8950 non-null   float64
 11  CASH_ADVANCE_TRX                  8950 non-null   int64  
 12  PURCHASES_TRX                     8950 non-null   int64  
 13  CREDIT_LIMIT                      8949 non-null   float64
 14  PAYMENTS                          8950 non-null   float64
 15  MINIMUM_PAYMENTS                  8637 non-null   float64
 16  PRC_FULL_PAYMENT                  8950 non-null   float64
 17  TENURE                            8950 non-null   int64  
dtypes: float64(14), int64(3), object(1)
memory usage: 1.2+ MB
In [6]:
creditcard_df.describe()
Out[6]:
BALANCE BALANCE_FREQUENCY PURCHASES ONEOFF_PURCHASES INSTALLMENTS_PURCHASES CASH_ADVANCE PURCHASES_FREQUENCY ONEOFF_PURCHASES_FREQUENCY PURCHASES_INSTALLMENTS_FREQUENCY CASH_ADVANCE_FREQUENCY CASH_ADVANCE_TRX PURCHASES_TRX CREDIT_LIMIT PAYMENTS MINIMUM_PAYMENTS PRC_FULL_PAYMENT TENURE
count 8950.000000 8950.000000 8950.000000 8950.000000 8950.000000 8950.000000 8950.000000 8950.000000 8950.000000 8950.000000 8950.000000 8950.000000 8949.000000 8950.000000 8637.000000 8950.000000 8950.000000
mean 1564.474828 0.877271 1003.204834 592.437371 411.067645 978.871112 0.490351 0.202458 0.364437 0.135144 3.248827 14.709832 4494.449450 1733.143852 864.206542 0.153715 11.517318
std 2081.531879 0.236904 2136.634782 1659.887917 904.338115 2097.163877 0.401371 0.298336 0.397448 0.200121 6.824647 24.857649 3638.815725 2895.063757 2372.446607 0.292499 1.338331
min 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 50.000000 0.000000 0.019163 0.000000 6.000000
25% 128.281915 0.888889 39.635000 0.000000 0.000000 0.000000 0.083333 0.000000 0.000000 0.000000 0.000000 1.000000 1600.000000 383.276166 169.123707 0.000000 12.000000
50% 873.385231 1.000000 361.280000 38.000000 89.000000 0.000000 0.500000 0.083333 0.166667 0.000000 0.000000 7.000000 3000.000000 856.901546 312.343947 0.000000 12.000000
75% 2054.140036 1.000000 1110.130000 577.405000 468.637500 1113.821139 0.916667 0.300000 0.750000 0.222222 4.000000 17.000000 6500.000000 1901.134317 825.485459 0.142857 12.000000
max 19043.138560 1.000000 49039.570000 40761.250000 22500.000000 47137.211760 1.000000 1.000000 1.000000 1.500000 123.000000 358.000000 30000.000000 50721.483360 76406.207520 1.000000 12.000000
In [7]:
# Check who made one off purchase of $40761 ie maximum ONEOFF_PURCHASES 
creditcard_df[creditcard_df['ONEOFF_PURCHASES'] == 40761.25]
Out[7]:
CUST_ID BALANCE BALANCE_FREQUENCY PURCHASES ONEOFF_PURCHASES INSTALLMENTS_PURCHASES CASH_ADVANCE PURCHASES_FREQUENCY ONEOFF_PURCHASES_FREQUENCY PURCHASES_INSTALLMENTS_FREQUENCY CASH_ADVANCE_FREQUENCY CASH_ADVANCE_TRX PURCHASES_TRX CREDIT_LIMIT PAYMENTS MINIMUM_PAYMENTS PRC_FULL_PAYMENT TENURE
550 C10574 11547.52001 1.0 49039.57 40761.25 8278.32 558.166886 1.0 1.0 0.916667 0.083333 1 101 22500.0 46930.59824 2974.069421 0.25 12
In [8]:
creditcard_df['CASH_ADVANCE'].max()
Out[8]:
47137.21176
In [9]:
# Check who made cash advance of $47137
# This customer made 123 cash advance transactions
# Never paid credit card in full
creditcard_df[creditcard_df['CASH_ADVANCE'] == 47137.211760000006]
Out[9]:
CUST_ID BALANCE BALANCE_FREQUENCY PURCHASES ONEOFF_PURCHASES INSTALLMENTS_PURCHASES CASH_ADVANCE PURCHASES_FREQUENCY ONEOFF_PURCHASES_FREQUENCY PURCHASES_INSTALLMENTS_FREQUENCY CASH_ADVANCE_FREQUENCY CASH_ADVANCE_TRX PURCHASES_TRX CREDIT_LIMIT PAYMENTS MINIMUM_PAYMENTS PRC_FULL_PAYMENT TENURE
In [10]:
# Check for missing Data 
creditcard_df.isnull().sum()
Out[10]:
CUST_ID                               0
BALANCE                               0
BALANCE_FREQUENCY                     0
PURCHASES                             0
ONEOFF_PURCHASES                      0
INSTALLMENTS_PURCHASES                0
CASH_ADVANCE                          0
PURCHASES_FREQUENCY                   0
ONEOFF_PURCHASES_FREQUENCY            0
PURCHASES_INSTALLMENTS_FREQUENCY      0
CASH_ADVANCE_FREQUENCY                0
CASH_ADVANCE_TRX                      0
PURCHASES_TRX                         0
CREDIT_LIMIT                          1
PAYMENTS                              0
MINIMUM_PAYMENTS                    313
PRC_FULL_PAYMENT                      0
TENURE                                0
dtype: int64
In [11]:
# Heat map for missing data  
sns.heatmap(creditcard_df.isnull(), yticklabels = False, cbar = False, cmap="Blues")
Out[11]:
<Axes: >
No description has been provided for this image
In [12]:
# Fill up the missing elements with mean of the 'MINIMUM_PAYMENT' 
creditcard_df.loc[(creditcard_df['MINIMUM_PAYMENTS'].isnull() == True), 'MINIMUM_PAYMENTS'] = creditcard_df['MINIMUM_PAYMENTS'].mean()
In [13]:
# Fill up the missing elements with mean of the 'CREDIT_LIMIT' 
creditcard_df.loc[(creditcard_df['CREDIT_LIMIT'].isnull() == True), 'CREDIT_LIMIT'] = creditcard_df['CREDIT_LIMIT'].mean()
In [14]:
creditcard_df.duplicated().sum()
Out[14]:
0
In [15]:
creditcard_df.isnull().sum()
Out[15]:
CUST_ID                             0
BALANCE                             0
BALANCE_FREQUENCY                   0
PURCHASES                           0
ONEOFF_PURCHASES                    0
INSTALLMENTS_PURCHASES              0
CASH_ADVANCE                        0
PURCHASES_FREQUENCY                 0
ONEOFF_PURCHASES_FREQUENCY          0
PURCHASES_INSTALLMENTS_FREQUENCY    0
CASH_ADVANCE_FREQUENCY              0
CASH_ADVANCE_TRX                    0
PURCHASES_TRX                       0
CREDIT_LIMIT                        0
PAYMENTS                            0
MINIMUM_PAYMENTS                    0
PRC_FULL_PAYMENT                    0
TENURE                              0
dtype: int64
In [16]:
sns.heatmap(creditcard_df.isnull(), yticklabels = False, cbar = False, cmap="Blues")
Out[16]:
<Axes: >
No description has been provided for this image
In [17]:
# Check for duplicated entries in the data
creditcard_df.duplicated().sum()
Out[17]:
0
In [18]:
# Drop Customer ID since it has no meaning here 
creditcard_df.drop("CUST_ID", axis = 1, inplace= True)
display(creditcard_df)
BALANCE BALANCE_FREQUENCY PURCHASES ONEOFF_PURCHASES INSTALLMENTS_PURCHASES CASH_ADVANCE PURCHASES_FREQUENCY ONEOFF_PURCHASES_FREQUENCY PURCHASES_INSTALLMENTS_FREQUENCY CASH_ADVANCE_FREQUENCY CASH_ADVANCE_TRX PURCHASES_TRX CREDIT_LIMIT PAYMENTS MINIMUM_PAYMENTS PRC_FULL_PAYMENT TENURE
0 40.900749 0.818182 95.40 0.00 95.40 0.000000 0.166667 0.000000 0.083333 0.000000 0 2 1000.0 201.802084 139.509787 0.000000 12
1 3202.467416 0.909091 0.00 0.00 0.00 6442.945483 0.000000 0.000000 0.000000 0.250000 4 0 7000.0 4103.032597 1072.340217 0.222222 12
2 2495.148862 1.000000 773.17 773.17 0.00 0.000000 1.000000 1.000000 0.000000 0.000000 0 12 7500.0 622.066742 627.284787 0.000000 12
3 1666.670542 0.636364 1499.00 1499.00 0.00 205.788017 0.083333 0.083333 0.000000 0.083333 1 1 7500.0 0.000000 864.206542 0.000000 12
4 817.714335 1.000000 16.00 16.00 0.00 0.000000 0.083333 0.083333 0.000000 0.000000 0 1 1200.0 678.334763 244.791237 0.000000 12
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
8945 28.493517 1.000000 291.12 0.00 291.12 0.000000 1.000000 0.000000 0.833333 0.000000 0 6 1000.0 325.594462 48.886365 0.500000 6
8946 19.183215 1.000000 300.00 0.00 300.00 0.000000 1.000000 0.000000 0.833333 0.000000 0 6 1000.0 275.861322 864.206542 0.000000 6
8947 23.398673 0.833333 144.40 0.00 144.40 0.000000 0.833333 0.000000 0.666667 0.000000 0 5 1000.0 81.270775 82.418369 0.250000 6
8948 13.457564 0.833333 0.00 0.00 0.00 36.558778 0.000000 0.000000 0.000000 0.166667 2 0 500.0 52.549959 55.755628 0.250000 6
8949 372.708075 0.666667 1093.25 1093.25 0.00 127.040008 0.666667 0.666667 0.000000 0.333333 2 23 1200.0 63.165404 88.288956 0.000000 6

8950 rows × 17 columns

In [19]:
n = len(creditcard_df.columns)
display(n)

17
In [20]:
display (creditcard_df.columns)

Index(['BALANCE', 'BALANCE_FREQUENCY', 'PURCHASES', 'ONEOFF_PURCHASES',
       'INSTALLMENTS_PURCHASES', 'CASH_ADVANCE', 'PURCHASES_FREQUENCY',
       'ONEOFF_PURCHASES_FREQUENCY', 'PURCHASES_INSTALLMENTS_FREQUENCY',
       'CASH_ADVANCE_FREQUENCY', 'CASH_ADVANCE_TRX', 'PURCHASES_TRX',
       'CREDIT_LIMIT', 'PAYMENTS', 'MINIMUM_PAYMENTS', 'PRC_FULL_PAYMENT',
       'TENURE'],
      dtype='object')
In [21]:
# distplot combines the matplotlib.hist function with seaborn kdeplot()
# KDE Plot represents the Kernel Density Estimate
# KDE is used for visualizing the Probability Density of a continuous variable. 
# KDE demonstrates the probability density at different values in a continuous variable. 

# Mean of balance is $1500
# 'Balance_Frequency' for most customers is updated frequently ~1
# For 'PURCHASES_FREQUENCY', there are two distinct group of customers
# For 'ONEOFF_PURCHASES_FREQUENCY' and 'PURCHASES_INSTALLMENT_FREQUENCY' most users don't do one off puchases or installment purchases frequently 
# Very small number of customers pay their balance in full 'PRC_FULL_PAYMENT'~0
# Credit limit average is around $4500
# Most customers are ~11 years tenure

plt.figure(figsize=(20,80))
for i in range(len(creditcard_df.columns)):
  plt.subplot(17, 1, i+1)
  sns.distplot(creditcard_df[creditcard_df.columns[i]], kde_kws={"color": "b", "lw": 3, "label": "KDE"}, hist_kws={"color": "g"})
  plt.title(creditcard_df.columns[i])

plt.tight_layout()
No description has been provided for this image
In [22]:
sns.pairplot(creditcard_df)
# Correlation between 'PURCHASES' and ONEOFF_PURCHASES & INSTALMENT_PURCHASES 
# Trend between 'PURCHASES' and 'CREDIT_LIMIT' & 'PAYMENTS'
Out[22]:
<seaborn.axisgrid.PairGrid at 0x166322810>
No description has been provided for this image
In [23]:
correlations = creditcard_df.corr()
display (correlations )
BALANCE BALANCE_FREQUENCY PURCHASES ONEOFF_PURCHASES INSTALLMENTS_PURCHASES CASH_ADVANCE PURCHASES_FREQUENCY ONEOFF_PURCHASES_FREQUENCY PURCHASES_INSTALLMENTS_FREQUENCY CASH_ADVANCE_FREQUENCY CASH_ADVANCE_TRX PURCHASES_TRX CREDIT_LIMIT PAYMENTS MINIMUM_PAYMENTS PRC_FULL_PAYMENT TENURE
BALANCE 1.000000 0.322412 0.181261 0.164350 0.126469 0.496692 -0.077944 0.073166 -0.063186 0.449218 0.385152 0.154338 0.531267 0.322802 0.394282 -0.318959 0.072692
BALANCE_FREQUENCY 0.322412 1.000000 0.133674 0.104323 0.124292 0.099388 0.229715 0.202415 0.176079 0.191873 0.141555 0.189626 0.095795 0.065008 0.114249 -0.095082 0.119776
PURCHASES 0.181261 0.133674 1.000000 0.916845 0.679896 -0.051474 0.393017 0.498430 0.315567 -0.120143 -0.067175 0.689561 0.356959 0.603264 0.093515 0.180379 0.086288
ONEOFF_PURCHASES 0.164350 0.104323 0.916845 1.000000 0.330622 -0.031326 0.264937 0.524891 0.127729 -0.082628 -0.046212 0.545523 0.319721 0.567292 0.048597 0.132763 0.064150
INSTALLMENTS_PURCHASES 0.126469 0.124292 0.679896 0.330622 1.000000 -0.064244 0.442418 0.214042 0.511351 -0.132318 -0.073999 0.628108 0.256496 0.384084 0.131687 0.182569 0.086143
CASH_ADVANCE 0.496692 0.099388 -0.051474 -0.031326 -0.064244 1.000000 -0.215507 -0.086754 -0.177070 0.628522 0.656498 -0.075850 0.303983 0.453238 0.139223 -0.152935 -0.068312
PURCHASES_FREQUENCY -0.077944 0.229715 0.393017 0.264937 0.442418 -0.215507 1.000000 0.501343 0.862934 -0.308478 -0.203478 0.568430 0.119778 0.103464 0.002976 0.305802 0.061506
ONEOFF_PURCHASES_FREQUENCY 0.073166 0.202415 0.498430 0.524891 0.214042 -0.086754 0.501343 1.000000 0.142329 -0.111716 -0.069088 0.544869 0.295030 0.243537 -0.029963 0.157531 0.082466
PURCHASES_INSTALLMENTS_FREQUENCY -0.063186 0.176079 0.315567 0.127729 0.511351 -0.177070 0.862934 0.142329 1.000000 -0.262958 -0.169207 0.529975 0.060752 0.085551 0.029590 0.250087 0.073275
CASH_ADVANCE_FREQUENCY 0.449218 0.191873 -0.120143 -0.082628 -0.132318 0.628522 -0.308478 -0.111716 -0.262958 1.000000 0.799561 -0.131168 0.132616 0.183192 0.097898 -0.249773 -0.133372
CASH_ADVANCE_TRX 0.385152 0.141555 -0.067175 -0.046212 -0.073999 0.656498 -0.203478 -0.069088 -0.169207 0.799561 1.000000 -0.066157 0.149699 0.255278 0.109185 -0.169784 -0.043421
PURCHASES_TRX 0.154338 0.189626 0.689561 0.545523 0.628108 -0.075850 0.568430 0.544869 0.529975 -0.131168 -0.066157 1.000000 0.272877 0.370832 0.095858 0.162066 0.121874
CREDIT_LIMIT 0.531267 0.095795 0.356959 0.319721 0.256496 0.303983 0.119778 0.295030 0.060752 0.132616 0.149699 0.272877 1.000000 0.421852 0.125134 0.055671 0.139034
PAYMENTS 0.322802 0.065008 0.603264 0.567292 0.384084 0.453238 0.103464 0.243537 0.085551 0.183192 0.255278 0.370832 0.421852 1.000000 0.125046 0.112138 0.106136
MINIMUM_PAYMENTS 0.394282 0.114249 0.093515 0.048597 0.131687 0.139223 0.002976 -0.029963 0.029590 0.097898 0.109185 0.095858 0.125134 0.125046 1.000000 -0.139674 0.057257
PRC_FULL_PAYMENT -0.318959 -0.095082 0.180379 0.132763 0.182569 -0.152935 0.305802 0.157531 0.250087 -0.249773 -0.169784 0.162066 0.055671 0.112138 -0.139674 1.000000 -0.016486
TENURE 0.072692 0.119776 0.086288 0.064150 0.086143 -0.068312 0.061506 0.082466 0.073275 -0.133372 -0.043421 0.121874 0.139034 0.106136 0.057257 -0.016486 1.000000
In [24]:
f, ax = plt.subplots(figsize = (20, 20))
sns.heatmap(correlations, annot = True)

# 'PURCHASES' have high correlation between one-off purchases, 'installment purchases, purchase transactions, credit limit and payments. 
# Strong Positive Correlation between 'PURCHASES_FREQUENCY' and 'PURCHASES_INSTALLMENT_FREQUENCY'
Out[24]:
<Axes: >
No description has been provided for this image
In [25]:
display (creditcard_df)
BALANCE BALANCE_FREQUENCY PURCHASES ONEOFF_PURCHASES INSTALLMENTS_PURCHASES CASH_ADVANCE PURCHASES_FREQUENCY ONEOFF_PURCHASES_FREQUENCY PURCHASES_INSTALLMENTS_FREQUENCY CASH_ADVANCE_FREQUENCY CASH_ADVANCE_TRX PURCHASES_TRX CREDIT_LIMIT PAYMENTS MINIMUM_PAYMENTS PRC_FULL_PAYMENT TENURE
0 40.900749 0.818182 95.40 0.00 95.40 0.000000 0.166667 0.000000 0.083333 0.000000 0 2 1000.0 201.802084 139.509787 0.000000 12
1 3202.467416 0.909091 0.00 0.00 0.00 6442.945483 0.000000 0.000000 0.000000 0.250000 4 0 7000.0 4103.032597 1072.340217 0.222222 12
2 2495.148862 1.000000 773.17 773.17 0.00 0.000000 1.000000 1.000000 0.000000 0.000000 0 12 7500.0 622.066742 627.284787 0.000000 12
3 1666.670542 0.636364 1499.00 1499.00 0.00 205.788017 0.083333 0.083333 0.000000 0.083333 1 1 7500.0 0.000000 864.206542 0.000000 12
4 817.714335 1.000000 16.00 16.00 0.00 0.000000 0.083333 0.083333 0.000000 0.000000 0 1 1200.0 678.334763 244.791237 0.000000 12
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
8945 28.493517 1.000000 291.12 0.00 291.12 0.000000 1.000000 0.000000 0.833333 0.000000 0 6 1000.0 325.594462 48.886365 0.500000 6
8946 19.183215 1.000000 300.00 0.00 300.00 0.000000 1.000000 0.000000 0.833333 0.000000 0 6 1000.0 275.861322 864.206542 0.000000 6
8947 23.398673 0.833333 144.40 0.00 144.40 0.000000 0.833333 0.000000 0.666667 0.000000 0 5 1000.0 81.270775 82.418369 0.250000 6
8948 13.457564 0.833333 0.00 0.00 0.00 36.558778 0.000000 0.000000 0.000000 0.166667 2 0 500.0 52.549959 55.755628 0.250000 6
8949 372.708075 0.666667 1093.25 1093.25 0.00 127.040008 0.666667 0.666667 0.000000 0.333333 2 23 1200.0 63.165404 88.288956 0.000000 6

8950 rows × 17 columns

In [26]:
# Let's scale the data first
scaler = StandardScaler()
creditcard_df_scaled = scaler.fit_transform(creditcard_df)
In [27]:
creditcard_df_scaled.shape
Out[27]:
(8950, 17)
In [28]:
display(pd.DataFrame(creditcard_df_scaled))
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
0 -0.731989 -0.249434 -0.424900 -0.356934 -0.349079 -0.466786 -0.806490 -0.678661 -0.707313 -0.675349 -0.476070 -0.511333 -0.960433 -0.528979 -3.109675e-01 -0.525551 0.360680
1 0.786961 0.134325 -0.469552 -0.356934 -0.454576 2.605605 -1.221758 -0.678661 -0.916995 0.573963 0.110074 -0.591796 0.688639 0.818642 8.931021e-02 0.234227 0.360680
2 0.447135 0.518084 -0.107668 0.108889 -0.454576 -0.466786 1.269843 2.673451 -0.916995 -0.675349 -0.476070 -0.109020 0.826062 -0.383805 -1.016632e-01 -0.525551 0.360680
3 0.049099 -1.016953 0.232058 0.546189 -0.454576 -0.368653 -1.014125 -0.399319 -0.916995 -0.258913 -0.329534 -0.551565 0.826062 -0.598688 4.878305e-17 -0.525551 0.360680
4 -0.358775 0.518084 -0.462063 -0.347294 -0.454576 -0.466786 -1.014125 -0.399319 -0.916995 -0.675349 -0.476070 -0.551565 -0.905464 -0.364368 -2.657913e-01 -0.525551 0.360680
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
8945 -0.737950 0.518084 -0.333293 -0.356934 -0.132643 -0.466786 1.269843 -0.678661 1.179833 -0.675349 -0.476070 -0.350408 -0.960433 -0.486217 -3.498541e-01 1.183951 -4.122768
8946 -0.742423 0.518084 -0.329136 -0.356934 -0.122823 -0.466786 1.269843 -0.678661 1.179833 -0.675349 -0.476070 -0.350408 -0.960433 -0.503396 4.878305e-17 -0.525551 -4.122768
8947 -0.740398 -0.185477 -0.401965 -0.356934 -0.294893 -0.466786 0.854576 -0.678661 0.760469 -0.675349 -0.476070 -0.390639 -0.960433 -0.570615 -3.354655e-01 0.329200 -4.122768
8948 -0.745174 -0.185477 -0.469552 -0.356934 -0.454576 -0.449352 -1.221758 -0.678661 -0.916995 0.157527 -0.182998 -0.591796 -1.097856 -0.580536 -3.469065e-01 0.329200 -4.122768
8949 -0.572575 -0.889033 0.042146 0.301732 -0.454576 -0.406205 0.439310 1.556082 -0.916995 0.990398 -0.182998 0.333524 -0.905464 -0.576869 -3.329464e-01 -0.525551 -4.122768

8950 rows × 17 columns

In [29]:
scores_1 = []

range_values = range(1, 20)

for i in range_values:
  kmeans = KMeans(n_clusters = i)
  kmeans.fit(creditcard_df_scaled)
  scores_1.append(kmeans.inertia_) 

plt.plot(scores_1, 'bx-')
plt.title('Finding the right number of clusters')
plt.xlabel('Clusters')
plt.ylabel('Scores') 
plt.show()

# From this we can observe that, 4th cluster seems to be forming the elbow of the curve. 
# However, the values does not reduce linearly until 8th cluster. 
# Let's choose the number of clusters to be 7.
No description has been provided for this image
In [30]:
kmeans = KMeans(8)
kmeans.fit(creditcard_df_scaled)
labels = kmeans.labels_
In [31]:
kmeans.cluster_centers_.shape
Out[31]:
(8, 17)
In [32]:
cluster_centers = pd.DataFrame(data = kmeans.cluster_centers_, columns = [creditcard_df.columns])
display(cluster_centers )
BALANCE BALANCE_FREQUENCY PURCHASES ONEOFF_PURCHASES INSTALLMENTS_PURCHASES CASH_ADVANCE PURCHASES_FREQUENCY ONEOFF_PURCHASES_FREQUENCY PURCHASES_INSTALLMENTS_FREQUENCY CASH_ADVANCE_FREQUENCY CASH_ADVANCE_TRX PURCHASES_TRX CREDIT_LIMIT PAYMENTS MINIMUM_PAYMENTS PRC_FULL_PAYMENT TENURE
0 -0.710404 -0.348568 -0.069657 -0.203109 0.208006 -0.444008 0.579687 -0.194433 0.582219 -0.641666 -0.455041 -0.032406 -0.049286 -0.206569 -0.297009 2.251728 0.130890
1 -0.226566 0.254336 -0.043341 -0.203397 0.271103 -0.335812 0.957736 -0.258514 1.131312 -0.407835 -0.321458 0.193606 -0.296811 -0.208180 0.086496 -0.306662 0.278313
2 2.058951 0.392258 -0.074157 -0.064124 -0.057840 2.958244 -0.214903 -0.054070 -0.162081 2.313287 2.736012 -0.059729 1.351496 1.513565 0.741851 -0.348590 -0.019274
3 1.526356 0.405003 7.356829 6.461232 5.521867 0.021953 1.075498 1.841044 1.035877 -0.314437 -0.112661 4.864643 2.242065 5.069151 1.204065 0.810717 0.330790
4 0.752829 0.404591 -0.365564 -0.254346 -0.397105 0.625632 -0.883494 -0.418778 -0.762496 1.102542 0.738791 -0.477212 0.238244 0.029338 0.238956 -0.442429 0.199499
5 -0.365914 -0.394812 -0.323888 -0.205536 -0.387993 -0.299175 -0.747326 -0.311164 -0.716239 -0.336099 -0.302707 -0.453405 -0.358020 -0.280702 -0.168230 -0.381427 0.273280
6 -0.343027 -0.469088 -0.291771 -0.207558 -0.308181 0.056474 -0.254356 -0.273245 -0.283325 0.288114 -0.014683 -0.396879 -0.559378 -0.397704 -0.202052 -0.097432 -3.200486
7 0.230723 0.451661 1.089843 1.027724 0.688834 -0.281540 1.131817 1.934360 0.618381 -0.373240 -0.294548 1.369933 0.777314 0.478822 -0.023544 0.329919 0.307844
In [33]:
# In order to understand what these numbers mean, perform inverse transformation
cluster_centers = scaler.inverse_transform(cluster_centers)
cluster_centers = pd.DataFrame(data = cluster_centers, columns = [creditcard_df.columns])
display(cluster_centers)

# First Customers cluster (Transactors): Those are customers who pay least amount of intrerest charges and careful with their money, Cluster with lowest balance ($104) and cash advance ($303), Percentage of full payment = 23%
# Second customers cluster (revolvers) who use credit card as a loan (most lucrative sector): highest balance ($5000) and cash advance (~$5000), low purchase frequency, high cash advance frequency (0.5), high cash advance transactions (16) and low percentage of full payment (3%)
# Third customer cluster (VIP/Prime): high credit limit $16K and highest percentage of full payment, target for increase credit limit and increase spending habits
# Fourth customer cluster (low tenure): these are customers with low tenure (7 years), low balanc
BALANCE BALANCE_FREQUENCY PURCHASES ONEOFF_PURCHASES INSTALLMENTS_PURCHASES CASH_ADVANCE PURCHASES_FREQUENCY ONEOFF_PURCHASES_FREQUENCY PURCHASES_INSTALLMENTS_FREQUENCY CASH_ADVANCE_FREQUENCY CASH_ADVANCE_TRX PURCHASES_TRX CREDIT_LIMIT PAYMENTS MINIMUM_PAYMENTS PRC_FULL_PAYMENT TENURE
0 85.827900 0.794698 854.381834 255.317927 599.165251 47.765379 0.723007 0.144455 0.595826 0.006740 0.143508 13.904328 4315.128564 1135.147894 172.039624 0.812306 11.692483
1 1092.896102 0.937521 910.605937 254.840409 656.222517 274.658079 0.874736 0.125338 0.814050 0.053532 1.055114 19.522159 3414.528811 1130.484192 1065.782352 0.064021 11.889773
2 5850.006716 0.970193 844.768305 486.004600 358.763705 7182.446638 0.404100 0.186328 0.300022 0.598057 21.920097 13.225182 9411.743341 6114.767624 2593.058885 0.051758 11.491525
3 4741.455043 0.973212 16721.182533 11316.758933 5404.423600 1024.907342 0.922000 0.751677 0.776121 0.072222 2.480000 135.626667 12652.000000 16407.838942 3670.229657 0.390835 11.960000
4 3131.424557 0.973115 222.171152 170.275114 51.970631 2290.849689 0.135762 0.077528 0.061402 0.355774 8.290536 2.848129 5361.277485 1818.075110 1421.083891 0.024312 11.784299
5 802.856116 0.783743 311.212183 251.289550 60.210203 351.486507 0.190413 0.109631 0.079786 0.067887 1.183073 3.439884 3191.827098 920.540430 472.154771 0.042154 11.883037
6 850.493066 0.766148 379.832326 247.933166 132.383813 1097.299957 0.388265 0.120943 0.251837 0.192799 3.148627 4.844911 2459.204819 581.829880 393.331757 0.125217 7.234249
7 2044.705236 0.984265 3331.670275 2298.249449 1033.971791 388.468841 0.944603 0.779515 0.610198 0.060455 1.238751 48.761249 7322.635445 3119.285237 809.337461 0.250210 11.929293
In [34]:
display(labels.shape) # Labels associated to each data point
display (labels.max())
display (labels.min())

(8950,)
7
0
In [35]:
y_kmeans = kmeans.fit_predict(creditcard_df_scaled)
display(y_kmeans)

array([0, 3, 7, ..., 2, 2, 2], dtype=int32)
In [36]:
# concatenate the clusters labels to our original dataframe
creditcard_df_cluster = pd.concat([creditcard_df, pd.DataFrame({'cluster':labels})], axis = 1)
creditcard_df_cluster.head()
Out[36]:
BALANCE BALANCE_FREQUENCY PURCHASES ONEOFF_PURCHASES INSTALLMENTS_PURCHASES CASH_ADVANCE PURCHASES_FREQUENCY ONEOFF_PURCHASES_FREQUENCY PURCHASES_INSTALLMENTS_FREQUENCY CASH_ADVANCE_FREQUENCY CASH_ADVANCE_TRX PURCHASES_TRX CREDIT_LIMIT PAYMENTS MINIMUM_PAYMENTS PRC_FULL_PAYMENT TENURE cluster
0 40.900749 0.818182 95.40 0.00 95.4 0.000000 0.166667 0.000000 0.083333 0.000000 0 2 1000.0 201.802084 139.509787 0.000000 12 5
1 3202.467416 0.909091 0.00 0.00 0.0 6442.945483 0.000000 0.000000 0.000000 0.250000 4 0 7000.0 4103.032597 1072.340217 0.222222 12 4
2 2495.148862 1.000000 773.17 773.17 0.0 0.000000 1.000000 1.000000 0.000000 0.000000 0 12 7500.0 622.066742 627.284787 0.000000 12 7
3 1666.670542 0.636364 1499.00 1499.00 0.0 205.788017 0.083333 0.083333 0.000000 0.083333 1 1 7500.0 0.000000 864.206542 0.000000 12 5
4 817.714335 1.000000 16.00 16.00 0.0 0.000000 0.083333 0.083333 0.000000 0.000000 0 1 1200.0 678.334763 244.791237 0.000000 12 5
In [37]:
for i in creditcard_df.columns:
  plt.figure(figsize = (20,5))
  for j in range(8):
    plt.subplot(1,8,j+1)
    cluster = creditcard_df_cluster[creditcard_df_cluster['cluster'] == j]
    cluster[i].hist(bins = 20)
    plt.title('{}    \nCluster {} '.format(i,j))
  
  plt.show()
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
In [38]:
pca = PCA(n_components=2)
principal_comp = pca.fit_transform(creditcard_df_scaled)
display(principal_comp)

array([[-1.68222026, -1.07645061],
       [-1.1382949 ,  2.50647666],
       [ 0.96968401, -0.3835203 ],
       ...,
       [-0.92620364, -1.8107856 ],
       [-2.33655167, -0.65796601],
       [-0.55642216, -0.40046712]])
In [39]:
# Create a dataframe with the two components
pca_df = pd.DataFrame(data = principal_comp, columns =['pca1','pca2'])
pca_df.head()
Out[39]:
pca1 pca2
0 -1.682220 -1.076451
1 -1.138295 2.506477
2 0.969684 -0.383520
3 -0.873628 0.043166
4 -1.599434 -0.688581
In [40]:
# Concatenate the clusters labels to the dataframe
pca_df = pd.concat([pca_df,pd.DataFrame({'cluster':labels})], axis = 1)
display(pca_df)
pca1 pca2 cluster
0 -1.682220 -1.076451 5
1 -1.138295 2.506477 4
2 0.969684 -0.383520 7
3 -0.873628 0.043166 5
4 -1.599434 -0.688581 5
... ... ... ...
8945 -0.359629 -2.016145 6
8946 -0.564369 -1.639123 6
8947 -0.926204 -1.810786 6
8948 -2.336552 -0.657966 6
8949 -0.556422 -0.400467 6

8950 rows × 3 columns

In [41]:
pca_df.value_counts(pca_df.cluster)
Out[41]:
cluster
5    2753
1    1763
4    1362
7    1085
0     879
6     619
2     414
3      75
Name: count, dtype: int64
In [42]:
plt.figure(figsize=(20,10))
ax = sns.scatterplot(x="pca1", y="pca2", hue = "cluster", data = pca_df, palette =['red','green','blue','pink','yellow','gray','purple', 'black'])
plt.show()
No description has been provided for this image
In [43]:
from tensorflow.keras.layers import Input, Add, Dense, Activation, ZeroPadding2D, BatchNormalization, Flatten, Conv2D, AveragePooling2D, MaxPooling2D, Dropout
from tensorflow.keras.models import Model, load_model
from tensorflow.keras.initializers import glorot_uniform  # This is normalizer
from keras.optimizers import SGD

encoding_dim = 7

input_df = Input(shape=(17,)) # 17 Features 


# Glorot normal initializer (Xavier normal initializer) draws samples from a truncated normal distribution 

x = Dense(encoding_dim, activation='relu')(input_df)
x = Dense(500, activation='relu', kernel_initializer = 'glorot_uniform')(x)
x = Dense(500, activation='relu', kernel_initializer = 'glorot_uniform')(x)
x = Dense(2000, activation='relu', kernel_initializer = 'glorot_uniform')(x)

encoded = Dense(10, activation='relu', kernel_initializer = 'glorot_uniform')(x)

x = Dense(2000, activation='relu', kernel_initializer = 'glorot_uniform')(encoded)
x = Dense(500, activation='relu', kernel_initializer = 'glorot_uniform')(x)

decoded = Dense(17, kernel_initializer = 'glorot_uniform')(x)

# autoencoder
autoencoder = Model(input_df, decoded)

#encoder - used for our dimention reduction
encoder = Model(input_df, encoded)

autoencoder.compile(optimizer= 'adam', loss='mean_squared_error')
In [44]:
display (creditcard_df_scaled.shape)

(8950, 17)
In [45]:
autoencoder.fit(creditcard_df_scaled, creditcard_df_scaled, batch_size = 128, epochs = 25,  verbose = 1)

Epoch 1/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 20ms/step - loss: 0.6548
Epoch 2/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 26ms/step - loss: 0.3745
Epoch 3/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 31ms/step - loss: 0.2987
Epoch 4/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 28ms/step - loss: 0.2637
Epoch 5/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 25ms/step - loss: 0.2259
Epoch 6/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 28ms/step - loss: 0.2042
Epoch 7/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 27ms/step - loss: 0.1839
Epoch 8/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 26ms/step - loss: 0.1660
Epoch 9/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 25ms/step - loss: 0.1554
Epoch 10/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 25ms/step - loss: 0.1668
Epoch 11/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 27ms/step - loss: 0.1363
Epoch 12/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 28ms/step - loss: 0.1361
Epoch 13/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 29ms/step - loss: 0.1347
Epoch 14/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 27ms/step - loss: 0.1237
Epoch 15/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 29ms/step - loss: 0.1189
Epoch 16/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 27ms/step - loss: 0.1157
Epoch 17/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 25ms/step - loss: 0.1028
Epoch 18/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 25ms/step - loss: 0.1033
Epoch 19/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 25ms/step - loss: 0.0963
Epoch 20/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 29ms/step - loss: 0.0998
Epoch 21/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 28ms/step - loss: 0.0975
Epoch 22/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 25ms/step - loss: 0.0882
Epoch 23/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 26ms/step - loss: 0.0910
Epoch 24/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 26ms/step - loss: 0.0828
Epoch 25/25
70/70 ━━━━━━━━━━━━━━━━━━━━ 2s 27ms/step - loss: 0.0810
Out[45]:
<keras.src.callbacks.history.History at 0x2adee6240>
In [46]:
autoencoder.summary()

Model: "functional"

┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ input_layer (InputLayer)        │ (None, 17)             │             0 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense (Dense)                   │ (None, 7)              │           126 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_1 (Dense)                 │ (None, 500)            │         4,000 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_2 (Dense)                 │ (None, 500)            │       250,500 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_3 (Dense)                 │ (None, 2000)           │     1,002,000 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_4 (Dense)                 │ (None, 10)             │        20,010 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_5 (Dense)                 │ (None, 2000)           │        22,000 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_6 (Dense)                 │ (None, 500)            │     1,000,500 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_7 (Dense)                 │ (None, 17)             │         8,517 │
└─────────────────────────────────┴────────────────────────┴───────────────┘

 Total params: 6,922,961 (26.41 MB)

 Trainable params: 2,307,653 (8.80 MB)

 Non-trainable params: 0 (0.00 B)

 Optimizer params: 4,615,308 (17.61 MB)
In [52]:
pred = encoder.predict(creditcard_df_scaled)
display (pd.DataFrame(pred))

280/280 ━━━━━━━━━━━━━━━━━━━━ 0s 2ms/step
0 1 2 3 4 5 6 7 8 9
0 0.0 1.430730 0.0 0.0 0.529805 0.072232 0.412029 0.0 0.706223 0.0
1 0.0 2.012459 0.0 0.0 0.999295 1.332797 2.951261 0.0 0.000000 0.0
2 0.0 3.829622 0.0 0.0 0.405321 2.754941 0.750495 0.0 1.722407 0.0
3 0.0 2.267767 0.0 0.0 0.168042 0.845819 0.416191 0.0 0.376821 0.0
4 0.0 1.333349 0.0 0.0 0.462254 0.269638 0.583908 0.0 0.672038 0.0
... ... ... ... ... ... ... ... ... ... ...
8945 0.0 1.377972 0.0 0.0 4.348932 0.000000 1.522304 0.0 3.482508 0.0
8946 0.0 1.006089 0.0 0.0 3.135665 0.000000 1.397675 0.0 3.713872 0.0
8947 0.0 1.680523 0.0 0.0 3.097224 0.000000 1.873369 0.0 3.893394 0.0
8948 0.0 3.434419 0.0 0.0 3.554084 0.000000 3.134046 0.0 3.907337 0.0
8949 0.0 6.092276 0.0 0.0 5.141526 2.463204 4.314789 0.0 7.287122 0.0

8950 rows × 10 columns

In [53]:
pred.shape
Out[53]:
(8950, 10)
In [54]:
scores_2 = []

range_values = range(1, 20)

for i in range_values:
  kmeans = KMeans(n_clusters= i)
  kmeans.fit(pred)
  scores_2.append(kmeans.inertia_)

plt.plot(scores_2, 'bx-')
plt.title('Finding right number of clusters')
plt.xlabel('Clusters')
plt.ylabel('scores') 
plt.show()
No description has been provided for this image
In [55]:
plt.plot(scores_1, 'bx-', color = 'r')
plt.plot(scores_2, 'bx-', color = 'g')
Out[55]:
[<matplotlib.lines.Line2D at 0x1402b6e40>]
No description has been provided for this image
In [56]:
kmeans = KMeans(4)
kmeans.fit(pred)
labels = kmeans.labels_
y_kmeans = kmeans.fit_predict(creditcard_df_scaled)
In [57]:
display(labels.shape) # Labels associated to each data point
display (labels.max())
display (labels.min())

(8950,)
3
0
In [58]:
y_kmeans = kmeans.fit_predict(pred)
display(y_kmeans)

array([0, 1, 1, ..., 1, 2, 2], dtype=int32)
In [59]:
df_cluster_dr = pd.concat([creditcard_df, pd.DataFrame({'cluster':labels})], axis = 1)
display(df_cluster_dr.head())
BALANCE BALANCE_FREQUENCY PURCHASES ONEOFF_PURCHASES INSTALLMENTS_PURCHASES CASH_ADVANCE PURCHASES_FREQUENCY ONEOFF_PURCHASES_FREQUENCY PURCHASES_INSTALLMENTS_FREQUENCY CASH_ADVANCE_FREQUENCY CASH_ADVANCE_TRX PURCHASES_TRX CREDIT_LIMIT PAYMENTS MINIMUM_PAYMENTS PRC_FULL_PAYMENT TENURE cluster
0 40.900749 0.818182 95.40 0.00 95.4 0.000000 0.166667 0.000000 0.083333 0.000000 0 2 1000.0 201.802084 139.509787 0.000000 12 1
1 3202.467416 0.909091 0.00 0.00 0.0 6442.945483 0.000000 0.000000 0.000000 0.250000 4 0 7000.0 4103.032597 1072.340217 0.222222 12 1
2 2495.148862 1.000000 773.17 773.17 0.0 0.000000 1.000000 1.000000 0.000000 0.000000 0 12 7500.0 622.066742 627.284787 0.000000 12 2
3 1666.670542 0.636364 1499.00 1499.00 0.0 205.788017 0.083333 0.083333 0.000000 0.083333 1 1 7500.0 0.000000 864.206542 0.000000 12 1
4 817.714335 1.000000 16.00 16.00 0.0 0.000000 0.083333 0.083333 0.000000 0.000000 0 1 1200.0 678.334763 244.791237 0.000000 12 1
In [60]:
pca = PCA(n_components=2)
prin_comp = pca.fit_transform(pred)
pca_df = pd.DataFrame(data = prin_comp, columns =['pca1','pca2'])
display (pca_df.head())
pca1 pca2
0 -1.917703 0.101947
1 0.010588 1.234658
2 0.991439 -1.104011
3 -1.448375 -0.136880
4 -1.835361 0.073406
In [61]:
pca_df = pd.concat([pca_df,pd.DataFrame({'cluster':labels})], axis = 1)
pca_df.head()
Out[61]:
pca1 pca2 cluster
0 -1.917703 0.101947 1
1 0.010588 1.234658 1
2 0.991439 -1.104011 2
3 -1.448375 -0.136880 1
4 -1.835361 0.073406 1
In [62]:
pca_df.value_counts(pca_df.cluster)
Out[62]:
cluster
1    5281
2    2289
3     973
0     407
Name: count, dtype: int64
In [63]:
plt.figure(figsize=(10,10))
ax = sns.scatterplot(x="pca1", y="pca2", hue = "cluster", data = pca_df, palette =['red','green','blue','yellow'])
plt.show()
No description has been provided for this image
In [ ]: